Understanding Structure Of Evil
Most people think of evil as individual bad acts. That's wrong. Evil tends to organize itself in patterns — structures that outlive any single person who participates in them. These structures reward cruelty or indifference without requiring anyone to feel personally cruel. Structure Of Evil refers to the way harmful outcomes become embedded in organizations, systems, and institutions. It's not about demons or malice. It's about design. A policy, a workflow, a set of incentives — that's where it lives. People don't have to want harm to produce it. They just have to follow the structure. The concept comes from philosophers like Hannah Arendt, who wrote about the banality of evil. She wasn't saying evil is ordinary. She was saying it runs through normal channels — paperwork, promotions, obedience to hierarchy. The structure does the work.
How It Actually Works In Practice
I've seen this play out in companies more times than I care to count. A clear example is when a business measures performance in a way that quietly encourages bad behavior. A call center that rates agents solely on call speed? Agents will rush customers off the phone. Nobody at the top meant to create a place where people feel unheard. The metric did that. The structure did the evil. Here's what beginners miss. They look for a villain. There usually isn't one. The problem is distributed. Twenty people each made small, reasonable decisions. The outcome was awful. Pinning it on anyone individually doesn't fix anything. Another counter-intuitive point: good people are the primary fuel for these structures. Ruthless people rarely stick around in places that need constant self-justification. They get uncomfortable. The ones who stay are the ones who believe they're doing fine work. That belief keeps the machine running.
Common Structural Patterns
Fragmentation of responsibility. When a task gets broken into too many pieces, no single person feels accountable for the final outcome. I worked on a project once where compliance, engineering, and product management all signed off on a feature. Each team checked its own box. The feature went live and caused real harm to users. Nobody had reviewed the combined effect. By the time anyone noticed, the damage was done. Incentive misalignment. Reward the wrong behavior and you'll get the wrong results. This is almost mechanical. A sales team compensated only on new bookings will push contracts people shouldn't sign. The compensation plan is the structure. The salespeople are just responding to it. Distance from consequences. The further someone is from the impact of their decisions, the less moral friction they experience. Middle managers often sit right in this gap. They see numbers, not people. The structure insulates them. That insulation is the problem.
Get the Full Details

Normalization. Once a harmful practice becomes routine, it stops being questioned. People stop seeing it. I remember a quarterly review where someone pointed out that our retention strategy involved not cancelling subscriptions even when customers clearly wanted out. The room went quiet. Everyone knew it was happening. Nobody had said it out loud until then. That silence is structure working.
How To Identify It
Start by mapping incentives. Write down who gets rewarded and for what. Then trace the path from those rewards to the actual outcomes on the ground. If there's a gap — and there usually is — that's where Structure Of Evil hides. Look for routines that have stopped being questioned. Anything your team does "because that's how it's always been done" deserves a hard look. Ask who benefits and who gets hurt. If the answer requires several layers of reasoning, you're probably looking at a structural problem. Check for information bottlenecks. When critical information only flows upward and never back down, the people making decisions are operating blind. The structure guarantees poor outcomes. This happens constantly in medium-to-large organizations.
How To Dismantle It
The hardest part is accepting that you're probably inside one of these structures right now. Not because you're bad. Because it's nearly impossible to build something large without creating some. The goal isn't perfection. It's awareness and constant adjustment. Create feedback loops. The people closest to the consequences need a direct line back to the people making decisions. Not through reports. Through actual conversation. I learned this the hard way when our support team flagged a recurring issue and the engineering roadmap didn't change for six months. The structure had silenced them. Once we set up biweekly direct sessions between support and product leads, things improved dramatically. Redistribute accountability. No single person should own a decision whose effects span multiple domains. Cross-functional review panels work better than individual sign-offs. They force people to confront the combined impact of their choices.

Rotate roles. People who only ever see one part of a system can't understand the whole. Job rotation — even informal shadowing — breaks the insulation. I had a manager who spent a month doing customer support tickets before taking on a product role. She changed how she designed features after that. Not because she became morally superior. Because she finally understood what her decisions looked like from the other side. Measure the right things. This sounds obvious and nobody actually does it well. Track outcomes, not activities. Track the impact on the people affected by your decisions, not the efficiency of the people making them. A metric that says "we shipped fast" is meaningless if the thing you shipped caused problems. Track "did the thing we shipped make things better or worse for the people using it?"
The Limits Of This Approach
Here's the honest part: you can't eliminate Structure Of Evil. You can only reduce it. Any organization large enough to matter will always produce unintended harmful outcomes. The best you can do is build mechanisms that surface those outcomes quickly and correct them. Some structures are harder to change than others. Government agencies, healthcare systems, financial institutions — the bigger and more regulated the environment, the more entrenched the patterns become. In those cases, the work is slower and the wins are smaller. Don't expect transformation. Expect incremental pressure. There's also a risk of what I call structural whack-a-mole. Fix one harmful pattern and another takes its place. This isn't failure. It's the nature of complex systems. The alternative — doing nothing — guarantees the harm continues unchanged.
If you're dealing with a structure that's causing serious harm right now, the fastest path isn't to redesign everything. It's to identify the single most impactful feedback loop and strengthen it. One real channel for ground-level information to reach decision-makers will do more than a dozen committee meetings about ethics. Structure Of Evil isn't something you defeat. It's something you manage. The people who treat it like a puzzle to be solved eventually burn out. The ones who treat it like a garden — constant weeding, regular pruning — tend to last longer and actually change things.
